4-Hydroxy-3-methoxymethamphetamine glucuronide as a phase II metabolite of 3,4-methylenedioxymethamphetamine: Enzyme-assisted synthesis and involvement of human hepatic uridine 5′-diphosphate-glucuronosyltransferase 2B15 in the glucuronidation

Article abstract of DOI:10.1248/cpb.57.472

3,4-Methylenedioxymethamphetamine (MDMA), one of the most popular illicit recreational drugs, is metabolized primarily into 4-hydroxy-3- methoxymethamphetamine (HMMA) by drug-metabolizing enzymes. HMMA is further metabolized by phase II enzymes to give th

Full text of DOI:10.1248/cpb.57.472

472
Chem. Pharm. Bull. 57(5) 472—475 (2009)
Vol. 57, No. 5
4-Hydroxy-3-methoxymethamphetamine Glucuronide as a Phase II
Metabolite of 3,4-Methylenedioxymethamphetamine: Enzyme-Assisted
ꢀ
Synthesis and Involvement of Human Hepatic Uridine 5 -Diphosphate-
Glucuronosyltransferase 2B15 in the Glucuronidation
*
Takuji SHODA, Kiyoshi FUKUHARA, Yukihiro GODA, and Haruhiro OKUDA
National Institute of Health Sciences; 1–18–1 Kamiyoga, Setagaya-ku, Tokyo 158–8501, Japan.
Received October 23, 2008; accepted February 10, 2009; published online February 12, 2009
3,4-Methylenedioxymethamphetamine (MDMA), one of the most popular illicit recreational drugs, is
metabolized primarily into 4-hydroxy-3-methoxymethamphetamine (HMMA) by drug-metabolizing enzymes.
HMMA is further metabolized by phase II enzymes to give the glucuronide or sulfate which is excreted into
urine. In the present study, enzyme kinetic studies with various microsomes showed that rat liver microsomes
pretreated with Aroclor 1254 were most suitable for the enzyme-assisted synthesis of the glucuronide (HMMA-
Gluc). This method selectively produced the b-anomer of HMMA-Gluc in a very high, isolated yield (71%), and
with a purity that was sufficient for use in an analysis of MDMA intake and for enzyme kinetic studies. We also
identified, by an LC-MS method, the human uridine 5ꢀ-diphosphate-glucuronosyltransferase (UGT) isoforms
that catalyze the glucuronidation of HMMA. Among 12 isoforms of human recombinant UGT expressed in
insect cells, UGT2B15 was the only isoform that showed adequate enzymatic activity in catalyzing HMMA
glucuronidation with Km and Vmax values of 3.8 mM and 1.6 nmol/min/mg protein, respectively. The finding that
UGT2B15 is capable of HMMA glucuronidation suggests this isoform may have an important in vivo role in
human MDMA metabolism.
Key words 3,4-methylenedioxymethamphetamine; enzyme-assisted synthesis; glucuronide; uridine 5ꢀ-diphosphate-glucurono-
syltransferase
The abuse of amphetamine-like designer drugs, which at synthetic HMMA-Gluc, we investigated the glucuronidation
present is a serious public health problem, is increasing activities of human uridine 5ꢀ-diphosphate-glucuronosyl-
among young adults. Although originally examined as an transferase (UGT) isoforms toward HMMA. In a series of
adjunct in psychotherapy, 3,4-methylenedioxymethampheta- UGT1A and 2B isoforms, UGT2B15 was found to catalyze
mine (MDMA), known in drug slang as “Ecstasy,” has re-
cently become the most popular recreational drug world-
wide.1—3)
The metabolic pathways of MDMA have been elucidated
and over a dozen metabolites have been identified in animals
and humans. The major MDMA metabolism pathway in-
volves demethylation to 3,4-dihydroxymethamphetamine
(DHMA) followed by O-methylation to 4-hydroxy-3-
methoxymethamphetamine (HMMA) as shown in Chart 1.
For unequivocal proof of MDMA use, detection of these
metabolites along with MDMA from a urine sample analysis
is required. HMMA exists as a major metabolite in urine,4—8)
whereas only a few studies concerning MDMA conjugates as
phase II metabolites have been reported. The HMMA level in
urine has been reported to increase after acid/enzymatic hy-
drolysis, suggesting that the majority of MDMA metabolites
are excreted into urine as the glucuronide or sulfate of
HMMA.9) If these conjugates were more readily available,
they could be used as analytical standards to obtain further
unequivocal proof of MDMA use. The synthesis of HMMA
glucuronide (HMMA-Gluc) was reported by Shima et al.,10)
but the yield by chemical synthesis was only 6%. For analy-
sis of MDMA metabolites and for enzyme kinetic studies of
the glucuronidation process, a more practical method to
prepare HMMA-Gluc is needed. In this study, we describe
the first enzyme-assisted synthesis of HMMA-Gluc using
Aroclor 1254-induced rat liver microsomes. This method
was easily performed and stereoselectively produced the b-
Chart 1. Principal Metabolic Pathway of MDMA in Humans
© 2009 Pharmaceutical Society of Japan
anomer conjugate in very high yield. In addition, using the
∗ To whom correspondence should be addressed. e-mail: tsho@nihs.go.jp

May 2009
473
Assay for HMMA Glucuronidation Using Microsomes Glucuronida-
tion of HMMA by three kinds of microsomes was determined by HPLC.
The incubation mixture (50 ml) contained 50 mM Tris–HCl (pH 7.5), 8 mM
MgCl2, 25 mg/ml alamethicin, 20 mM UDPGA, microsomes and HMMA.
the glucuronidation of HMMA.
Experimental
General Information Uridine 5ꢀ-diphosphoglucuronic acid (UDPGA)
and alamethicin were purchased from Sigma-Aldrich (St. Louis, MO, Concentrations of microsomal proteins were 0.48 mg protein/ml for Aroclor
U.S.A.). All other reagents and solvents were purchased from Wako Pure 1254-induced rat liver microsomes, 0.48 mg protein/ml for non-induced rat
Chemical (Osaka, Japan), Tokyo Kasei Kogyo (Tokyo, Japan), and Kanto liver microsomes, and 0.5 mg protein/ml for human liver microsomes.
Chemical (Tokyo, Japan), and were used without purification. Male
HMMA concentrations varied from 1.0 to 40 mM. Incubations were per-
Sprague-Dawley rat liver microsomes (Lot No. JJS) and Aroclor 1254-in- formed in a water bath at 37 °C, for 10 min with Aroclor 1254-induced and
duced Male Sprague-Dawley rat liver microsomes (Lot No. ADM) were pur- non-induced rat liver microsomes, and for 30 min with human liver micro-
chased from Charles River Laboratories (Wilmington, MA, U.S.A.). Pooled somes. The enzyme assays were terminated by the addition of 50 ml of 10%
human liver microsomes (Lot No. 70196) and microsomes from baculo- HClO4, briefly vortexed, and then centrifuged at 3000 rpm and 4 °C for
virus-insect cells expressing UGTs 1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 10 min. The supernatants were injected into the HPLC. The absorbance of
1A10, 2B4, 2B7, 2B15, 2B17 were purchased from BD Gentest (Woburn,
HMMA-Gluc at 275 nm was detected, and the peak area was determined for
MA, U.S.A.). Control experiments were carried out using microsomes kinetic analysis. The apparent Km and Vmax were estimated by analyzing
from insect cells infected with wild type baculovirus purchased from BD
Gentest.
Michaelis–Menten plots using KaleidaGraph ver. 4.0 software (Synergy
Software, Reading, PA, U.S.A.).
Analytical HPLC was performed using a CBM-20A system controller,
LC-20A pump, SPD-M20A UV/Vis photodiode array detector, and CTO-
10AC column oven (Shimadzu, Kyoto, Japan) equipped with a CAPCELL
UGT Assay for HMMA Glucuronidation Enzyme assay mixtures
(50 ml final volume) contained 50 mM Tris–HCl (pH 7.5), 8 mM MgCl2,
25 mg/ml alamethicin, 5 mM UDPGA, 5 mM HMMA, and 1 mg protein/ml of
PAK C18 MGII 5 mm, 4.6ꢁ250 mm (Shiseido, Tokyo, Japan). The mobile microsomes expressing UGT isoforms or control microsomes. Incubation
phases were A: 0.1% trifluoroacetic acid (TFA)/H2O and B: 0.1%
was performed for 120 min at 37 °C and was terminated by the addition of
TFA/CH3CN. Preparative HPLC was performed using a SSC-6600 gradient 50 ml of 10% HClO4. The mixture was briefly vortexed, then centrifuged at
controller, SSC-3465 pump, SSC-5410 UV/Vis detector and SSC-3465 col-
3000 rpm and 4 °C for 10 min. The supernatants were analyzed by LC-MS.
umn oven equipped with a SenshuPak PEGASIL ODS column, 5 mm The mass spectrometer was operated in the selected ion monitoring mode
20ꢁ250 mm (Senshu Kagaku, Tokyo, Japan). The mobile phases were A: using the [MꢄH]ꢄ ion, m/z 372, for HMMA-Gluc.
0.1% TFA/H2O and B: 0.1% TFA/CH3CN. LC-MS was performed using a
Dual l Absorbance Detector 2487, micromass ZQ and an Alliance model
The kinetic parameters for UGT2B15 processing of HMMA were esti-
mated by analyzing Michaelis-Menten plots using KaleidaGraph ver. 4.0
2695 (Waters, Milford, MA, U.S.A.) equipped with a CAPCELL PAK C18 software.
MGII 5 mm, 4.6ꢁ250 mm (Shiseido). The mobile phases were composed of
A: 0.1% HCOOH/H2O and B: CH3CN. 1H- and 13C-NMR spectra were
Results
1
recorded on a Varian AS 400 Mercury spectrometer (400 MHz for H and
Enzyme-Assisted Synthesis of HMMA-Gluc In the
100 MHz for 13C). Chemical shifts are expressed in ppm downfield from
sodium 3-(trimethylsilyl)-propionate-2,2,3,3,-d4 (d scale). High resolution
mass spectra were obtained on a LTQ Orbitrap (Thermo Fischer Scientific,
Waltham, MA, U.S.A.).
case of catechol derivatives, large scale (mg) syntheses of
their glucuronides have been accomplished by an enzyme-
associated method using rat liver microsomes pretreated
with Aroclor 1254, a mixture of polychlorinated biphenyls,
as biocatalyst.13) We applied this method to the synthesis of
HMMA-Gluc because of the structural similarities between
HMMA and catechol. Concentrations of UDPGA and
HMMA were 5 mM and 10 mM, respectively and alamethicin,
a pore-forming peptide, was added to the reaction system to
activate UGT in the liver microsomes without affecting CYP
activity.14)
Synthesis of HMMA The synthesis of HMMA was carried out accord-
ing to previous reports,11,12) with slight modifications. A solution of 5.0 g of
4-hydroxy-3-methoxyphenylacetone in anhydrous methanol (50 ml) was
added to a solution of 12 ml (117 mmol) of 40% methylamine in methanol
solution, followed by the addition of 2.65 g (42 mmol) of NaBH3CN in sev-
eral portion. The solution was adjusted to pH 6 with 2.0 M HCl and stirred at
room temperature for 24 h under N2. The mixture was then poured into
100 ml of water, adjusted to pH 2 with concentrated HCl, and stirred at room
temperature for 1 h. After evaporation to remove methanol, the aqueous
layer was washed with ether, basicified by the portionwise addition of NaOH
until pHꢂ12, and extracted twice with ether. The organic layer was dried
over MgSO4, concentrated in vacuo, and 2.0 M HCl in ether was added until
efficient precipitation was observed. The precipitate was collected and re-
crystallized from ethanol to afford HMMA as a white solid (HCl salt, 3.44 g,
yield 53%). mp 217—218 °C, 1H-NMR (CD3OD) d: 1.01 (d, Jꢃ6.4 Hz, 3H),
A representative chromatogram of the reaction mixture is
shown in Fig. 1. During the course of the reaction, peak C at
7.2 min was found to increase with a concomitant decrease in
HMMA, peak D, at 9.2 min. The purification of peak C was
2.34 (s, 3H), 2.46 (dd, Jꢃ7.0, 12.8 Hz, 1H), 2.66—2.77 (m, 2H), 3.81 (s, therefore carried out by preparative HPLC after protein pre-
3H), 6.60 (dd, Jꢃ2.0, 8.0 Hz, 1H), 6.72—6.74 (m, 2H), 13C-NMR (CD3OD)
d: 18.9, 33.5, 43.3, 56.3, 57.7, 113.8, 116.4, 122.7, 131.4, 146.4, 149.1,
MS (ESI) 196 [MꢄH]ꢄ, 165 [MꢄHꢅNHCH3]ꢄ, HR-MS (ESI) Found
196.1334, Calcd for C11H18O2Nꢄ 196.1332.
cipitation. After elimination of the solvent, a white solid was
obtained. From analysis of H-NMR, 13C-NMR, and MS
1
spectra, the structure of the product was determined to be
HMMA-Gluc. The advantage of enzyme-assisted synthesis is
that the formation solely of biologically relevant stereo- and
Enzyme-Assisted Synthesis of HMMA-Gluc Ten milliliters of a buffer
solution containing 50 mM Tris–HCl (pH 7.5), 8 mM MgCl2, 25 mg/ml
alamethicin, 5 mM UDPGA, and 10 mM HMMA was stirred in a 37 °C water
bath, and the reaction was started by the addition of 200 ml of Aroclor 1254-
induced rat liver microsomes (24 mg protein/ml) and stirred continuously for
20 h. The reaction was stopped with 5 ml of 10% HClO4. The precipitated
proteins were removed by centrifugation (3000 rpm, 10 min, 4 °C), and the
supernatant was filtered. The filtrate was purified by preparative HPLC, and
after evaporation of the fraction containing the product, HMMG-Gluc was
obtained as a white solid (13.2 mg, yield 71% from UDPGA). 1H-NMR
(D2O) d: 1.27 (d, Jꢃ6.4 Hz, 3H), 2.69 (s, 3H), 2.87 (dd, Jꢃ7.2, 14.0 Hz,
1H), 3.00 (dd, Jꢃ7.2, 14.0 Hz, 1H), 3.49—3.54 (m, 1H), 3.62—3.69 (m,
3H), 3.88 (s, 3H), 3.99 (d, Jꢃ9.2 Hz, 1H), 5.14 (d, Jꢃ7.2 Hz, 1H), 6.86 (dd,
Jꢃ8.4, 2.0 Hz, 1H), 6.99 (d, Jꢃ2.0 Hz, 1H), 7.14 (d, Jꢃ8.4 Hz, 1H), 13C-
NMR (D2O) d: 15.0, 30.2, 38.5, 56.0, 56.5, 71.2, 72.6, 74.6, 75.2, 100.6,
114.0, 116.9, 122.3, 131.7, 144.5, 149.2, 172.0, MS, (ESI) 372 [MꢄH]ꢄ,
196 [MꢄH-glucro]ꢄ, 165 [MꢄH-glucro-NHCH3]ꢄ, 370 [MꢅH]ꢅ, HR-MS
(ESI) Found 372.1659, Calcd for C17H26O8Nꢄ 372.1653.
Fig. 1. HPLC Chromatogram after Enzymatic Synthesis
A and B, by-products derived from microsomes or UDPGA; C, HMMA-Gluc; D,
HMMA.

474
Vol. 57, No. 5
regioisomeric products can be expected. In fact, the stereo- dation. Therefore, enzyme analyses were performed on the
chemistry of the anomeric proton H1ꢀ, identified by its char- 12 kinds of human microsomes that express UGT isoforms
acteristic chemical shift at 5.15 ppm, was found by analysis (UGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4,
of the coupling constant between H1ꢀ and H2ꢀ (7.2 Hz) to be, 2B7, 2B15, 2B17). Initially, an HPLC method to quantitate
as expected, in the b-configuration. Although purely chemi- the amount of HMMA-Gluc in the enzyme reaction mixture
cal syntheses of HMMA-Gluc have been low yielding, this was tried; however, it was difficult to detect the UV ab-
enzyme-assisted synthesis was found to stereoselectively sorbance from HMMA-Gluc because of overlap from impu-
produce HMMA-Gluc in very high yield (71%).
rity peaks derived from insect cell microsomes (data not
Kinetic Analysis of Glucuronidation in Microsomes shown). Therefore, an LC-MS method, in selected ion moni-
UGT activities in microsomes are expected to differ with re- toring mode set for HMMA-Gluc at m/z 372, was used. Re-
spect to species and microsomal enzyme inducer. To study sults from the LC-MS study are shown in Fig. 2. In the case
the biocatalytic efficiency of various microsomes toward glu- of the UGT1A series, HMMA-Gluc was hardly detectable,
curonidation, kinetic parameters for HMMA glucuronidation even for the 1A1, 1A6 and 1A9 isozymes which are known
were determined using Aroclor 1254-induced and non-in- to catalyze the glucuronidation of hydroxyl groups in pheno-
duced rat liver and human liver microsomes. All three types lic compounds.15) In contrast, production of HMMA-Gluc
of microsomes catalyzed HMMA glucuronidation and the was observed in the UGT2B series, with UGT2B15 being
apparent kinetic parameters are shown in Table 1. The appar- especially productive. The Vmax and Km values for human
ent Vmax of Aroclor 1254-induced rat liver microsomes was UGT2B15 were determined to be 1.6 nmol/min/mg protein
64.7 nmol/min/mg protein, two fold higher than from non- and 3.8 mM, respectively, showing that human UGT2B15 ex-
induced rat liver microsomes. Human liver microsomes hibits high glucuronidation efficiency with HMMA.
showed a lower apparent Vmax as compared with both rat liver
microsomes. Km values of 1.9—7.6 mM were obtained for all Discussion
three microsomes which, while low, are sufficient for the
MDMA metabolism involves demethylation to DHMA
purpose of enzymatic synthesis. These results suggest that catalyzed by CYP2D6, followed by O-methylation to
Aroclor 1254-induced rat liver microsomes are a suitable HMMA, catalyzed by COMT. Although some HMMA is de-
biocatalyst for an enzyme-assisted synthesis of HMMA-Gluc. tected in urine samples as a metabolite of MDMA, most of it
Human UGT Isozyme Catalysis of HMMA Glu- is conjugated by phase II enzymes to give a glucuronide or
curonidation It is known that the metabolism of MDMA sulfate which is excreted into urine.9,10) If metabolites of
involves N-demethylation by CYP2D6 to DHMA. DHMA MDMA were readily available as authentic standards, the de-
can undergo subsequent O-methylation mediated by catechol tection of these metabolites along with HMMA or MDMA in
O-methyltransferase (COMT) to HMMA as shown in Chart urine samples would constitute a powerful analytical method
1. The hydroxyl group of HMMA can be further metabolized for unequivocal proof of MDMA use. Unfortunately, the con-
by phase II enzymes to produce conjugates, the glucuronide/ jugate metabolites are not commercially available and are
sulfate, which are excreted into urine. Although the majority difficult to synthesize. In this report, we focused on HMMA-
of MDMA metabolites are conjugates, there have been no re- Gluc as a major urinary metabolite of MDMA and developed
ports on the UGT isoforms involved in HMMA glucuroni- a useful method based on enzyme-assisted synthesis using
rat liver microsomes for its production. Liver microsomes
from Aroclor 1254-induced rats were used as a highly active
source of mammalian UGT. After purification by protein pre-
cipitation and preparative HPLC, the structure of HMMA-
Table 1. Kinetic Parameters for HMMA Glucuronidation by Liver Micro-
somes
Aroa)
Ratb)
Humanc)
1
Gluc was characterized by H-NMR and mass spectrometry.
This enzymatic method was highly stereoselective, producing
the b-anomer. Pure HMMA-Gluc was isolated in very high
yield (71%, 13.2 mg), in amounts sufficient for the analysis
of MDMA consumption and for enzyme kinetics studies.
An LC-MS method was employed for identification of
glucuronidation activities of human UGT isoforms toward
HMMA. High levels of enzymatic activity responsible for
the formation of HMMA-Gluc from HMMA were only ob-
served in UGT2B15 which is an isoform that catalyzes the
glucuronidation of flavonoids and androgenic and estrogenic
steroids.16) A number of endogenous phenolic compounds,
drugs, and hydroxylated metabolites of drugs are substrates
for the UGT1A series, but interestingly, HMMA is not glu-
curonidated by any of its members.
Vmax (nmol/min/mg protein)
Km (mM)
64.7
6.7
30.6
1.9
16.7
7.6
a) Aroclor 1254-induced rat liver microsomes. b) Non-induced rat liver micro-
somes. c) Pooled human liver microsomes.
Oxazepam, a commonly used 1,4-benzodiazepine anxio-
lytic drug, is known to be cleared primarily by glucuronida-
tion catalyzed by UGT2B15.17) The glucuronidation of ox-
azepam by human subjects is gender dimorphic, arising from
a genetic polymorphism for UGT2B15. Such an effect of
UGT2B15 polymorphism on gender differences may be con-
Fig. 2. Glucuronidation of HMMA by Microsomes from Insect Cells Ex-
pressing Human UGT Isoforms
∗ Not detected.

May 2009
475
(2003).
sidered in the context of MDMA activity, i.e., it is known
that increasing doses of MDMA produce more hallucinogen
like perceptual alterations, particularly in women. As in the
case of oxazepam, this gender difference might be attributa-
ble to the involvement of UGT2B15 in the glucuronidation of
HMMA.
In conclusion, we have demonstrated a large scale prepara-
tion of HMMA-Gluc by enzyme-assisted synthesis. Kinetic
parameters indicated that using Aroclor 1254-induced rat
liver microsomes was superior to using non-induced rat liver
microsomes or human liver microsomes. In addition, we
showed HMMA is glucuronidated to HMMA-Gluc by human
UGT2B15. Further studies, utilizing enzyme-assisted synthe-
sis of other illegal drug metabolites that are difficult to pre-
pare by chemical synthesis, for example, psilocin, are now in
progress.
4) Maurer H. H., Bickeboeller-Friedrich J., Kraemer T., Peters F. T., Toxi-
col. Lett., 112—113, 133—142 (2000).
5) Lim H. K., Foltz R. L., Chem. Res. Toxicol., 2, 142—143 (1989).
6) de la Torre R., Farre M., Roset P. N., Lopez C. H., Mas M., Ortuno J.,
Menoyo E., Pizarro N., Segura J., Cami J., Ann. N. Y. Acad. Sci., 914,
225—237 (2000).
7) de la Torre R., Farre M., Roset P. N., Pizarro N., Abanades S., Segura
M., Segura J., Cami J., Ther. Drug Monit., 26, 137—144 (2004).
8) Yamada H., Ishii Y., Oguri K., J. Health Sci., 51, 1—7 (2005).
9) Pirnay S. O., Abraham T. T., Lowe R. H., Huestis M. A., J. Anal. Toxi-
col., 30, 563—569 (2006).
10) Shima N., Kamata H., Katagi M., Tsuchihashi H., Sakuma T., Nemoto
N., J. Chromatogr. B, 857, 123—129 (2007).
11) Pizarro N., de la Torre R., Farre M., Segura J., Llebaria A., Joglar J.,
Bioorg. Med. Chem., 10, 1085—1092 (2002).
12) Morgan P. H., Beckett A. H., Tetrahedron, 31, 2595—2601 (1975).
13) Luukkanen L., Kilpelainen I., Kangas H., Ottoila P., Elovaara E., Task-
inen J., Bioconjug. Chem., 10, 150—154 (1999).
14) Fisher M. B., Campanela K., Ackermann B. L., Vandenbrandenn M.,
Wrighton S. A., Drug Metab. Dispos., 28, 560—566 (2000).
15) Tukey R. H., Strassburg C. P., Annu. Rev. Pharmacol. Toxicol., 40,
581—616 (2000).
Acknowledgements This work was supported by a Grant from the Min-
istry of Health, Labour and Welfare, Japan.
References
16) Green M. D., Oturu E. M., Tephly T. R., Drug Metab. Dispos., 22,
799—805 (1994).
1) Nichols D. E., J. Psychoactive Drugs, 18, 305—313 (1986).
2) Strote J., Lee J. E., Wechsler H., J. Adolesc. Health, 30, 64—72 17) Court M. H., Hao Q., Krishnaswamy S., Bekaii-Saab T., Al-Rohaimi
(2002).
A., von Moltke L. L., Greenblatt D. J., J. Pharmacol. Exp. Ther., 310,
3) Schifano F., Oyefeso A., Corkery J., Cobain K., Jambert-Gray R., Mar-
tinotti G., Ghodse A. H., Hum. Psychopharmacol., 18, 519—524
656—665 (2004).